Home Neurofeedback Devices Compared: Prices & Specs (2026)

The comparison table

Home neurofeedback runs from $379 for a consumer headband to $13,994 for a professional system. Clinician-supervised programs with a brain map sit between $3,999 and $9,499; a NeurOptimal rental is $650 to $799 a month. Four different technologies are sold under the word "neurofeedback," and they are not tiers of one product. The table gives the prices. The sections after it explain what separates them.

Prices verified 5 September 2026. LENS figures verified with Ochs Labs and BrainTap pricing taken from braintap.com, both 7 September 2026.

BrainTap is in the table because people compare it with neurofeedback devices when shopping, not because it is one. It plays light and sound at a target frequency; it does not measure brain activity, so there is no feedback loop. Its age range, user allowance and return terms were not checked for this update.

The last three rows are not devices you buy and operate. LENS is a brief, passive system administered by a trained provider. Peak Brain and Myneurva are clinician-supervised programs built around a QEEG brain map, with the training sessions done at home. They are here because leaving provider-directed options out would make the rest of the table look like the whole market.

Quick glance: start from what brought you here

Most comparison pages are organized around devices. People usually begin somewhere else — with attention, sleep, stress, or a body that keeps reacting after therapy has done its work. This table starts from those concerns and shows what people in each situation tend to consider.

Use this as orientation, not as a recommendation. Neurofeedback is brain training, not treatment. None of these systems diagnoses a condition or replaces medical or psychological care, and no comparison page can determine which one will work for you. A figure commonly repeated in practitioner literature is that 15 to 30 percent of people do not respond to neurofeedback whichever approach they use; it is an estimate rather than a finding from a single controlled trial, and it should be read that way. What this table can do is narrow the category and say what the research does and does not support.

First: four different products are sold as "neurofeedback"

The word covers four technologies that work in fundamentally different ways. Knowing which one you are looking at is the first thing to establish, because a price only means something once you know what it buys.

EEG neurofeedback measures electrical activity at the scalp and provides feedback based on that activity. Narbis, Myndlift, Neurosity, Sens.ai, protocol-based professional systems, NeurOptimal and LENS all use EEG, but not in the same way. Protocol-based systems reward a selected frequency or pattern. NeurOptimal is non-linear dynamical and briefly interrupts audio when it detects turbulence, without setting a target. Channel count alone does not capture these differences.

Low-energy electromagnetic feedback is what LENS does, and it is passive. It measures EEG and returns a very low-energy electromagnetic signal through the scalp sensors rather than using an audio or visual reward. There is no game to play and no state to produce deliberately.

fNIRS — functional near-infrared spectroscopy — measures blood oxygenation in the prefrontal cortex rather than electrical activity. Mendi uses fNIRS, while Muse S Athena combines fNIRS and EEG. fNIRS is a valid form of measurement, but it is not EEG, and training blood flow in one area of the forehead is narrower than monitoring electrical activity across multiple sites.

Audio-visual entrainment uses flashing lights and pulsing tones at a target frequency. It is not neurofeedback, because nothing is measured and there is no feedback loop. BrainTap and similar devices belong in this category. BrainTap sells the headset at $797, or $907 bundled with a year of app membership; membership after that is $29.99 a month or $260 a year. Some people find these devices useful, but they should be evaluated separately from EEG and fNIRS systems, because they are answering a different question.

If a product page does not clearly identify which of these four it uses, look for clarification before comparing it with anything else.  If you want the mechanism rather than the specs, we cover how neurofeedback works separately. 

Second: not all EEG is equal

Two devices can both be described as EEG neurofeedback and work in entirely different ways. Price gives you a rough signal about build quality, but it says almost nothing about how a system decides what to do with the signal it collects. Three questions separate them.

How is the signal picked up, and how does it travel?

Dry sensors on a headband make contact through the hair and skin without conductive paste. This makes consumer systems easier and less expensive to use at home. The tradeoff is a coarser signal that is generally more susceptible to movement and contact artifacts than sensors applied with a conductive medium.

The connection also matters, because neurofeedback is designed to deliver feedback while the brain activity that produced it is still occurring. Bluetooth can introduce latency and, on a congested connection, dropped packets. That may matter little when tracking a general trend across a session, such as whether a focus score improved. In real-time training, however, a delay can weaken the timing between the brain activity and the feedback. A wired connection removes that variable.

These differences affect signal quality and timing, but they do not mean that dry-sensor devices are ineffective.

What is done with the data once it is collected?

This question reveals some of the most important differences among systems.

Target-driven systems decide in advance what the brain should be doing — for example, producing more alpha or changing the theta-to-beta ratio — and reward movement toward that target. The target may be selected by a clinician or built into the manufacturer’s default protocol.

Information-driven systems do not set a target. They monitor the signal and give the brain information about its own activity. In NeurOptimal’s case, the audio is briefly interrupted when the system detects turbulence in the signal. Rather than directing the brain toward a predetermined state, the system allows it to respond to the information.

LENS is a separate design again. It identifies the dominant EEG frequency and briefly returns a very low-energy electromagnetic signal through the sensors. The client does not have to focus on a screen, control a game or consciously produce a particular state. A trained provider selects the sensor sites, sequence and exposure.

The range of brainwaves being monitored matters alongside this. A system training one frequency band at one site is watching a narrow slice. NeurOptimal is designed to respond across a wide frequency range in real time rather than to a single band. Wide-range monitoring and target-free design are engineering choices you can evaluate directly, whatever you conclude about them.

An important caveat

Design sophistication and research volume are different axes, and they do not point the same way. Target-driven protocol training is the most extensively studied form of neurofeedback — attention-related protocols in particular have decades of peer-reviewed research behind them, considerably more than newer approaches have accumulated. A more capable design is not automatically the better-evidenced one.

The independent evidence for LENS is limited. The small controlled trials published to date — in fibromyalgia, and in stress, anxiety and cognition — did not find significant differences between active LENS and sham conditions. That does not establish that LENS is ineffective, and it is a small literature that may move; it does mean broad outcome claims should be read cautiously.

The table below sets out what each system measures and what it is designed to do. Whether it produces meaningful changes depends on the individual.

Third: what each device says it does, and what its design supports

Each device has a specific design and intended use. The useful question is how closely its claims align with what the technology is designed to do. Read the right-hand column as a practical interpretation of the design.

Does price track capability?

To some extent, but mainly within comparable categories.

Within EEG neurofeedback, price can reflect the features described above. Higher-quality sensors, a wired connection, wider frequency coverage and professional-grade construction cost more to build, and those differences affect the quality and timing of the signal.

Across categories, however, price is a less reliable guide to capability. Neurosity Crown costs $1,499 and functions primarily as a measurement and development platform; it is not necessarily more clinically oriented than the $690 Narbis. Muse at $474.99 and Narbis at $690 are not two levels of the same product — they are designed for different purposes. Myndlift’s $150 monthly fee includes clinician oversight rather than higher-specification hardware. Renting NeurOptimal for $650–799 per month provides the same system sold for $5,995–9,995, with a different payment structure. The $13,994 LENS package is professional equipment that includes training and a configured system; it is not a consumer purchase comparable to a headband.

The more useful comparison is within a category. First decide whether you want a wellness tool, a single-protocol trainer, a supervised clinical program, or professional-grade equipment available for home use. Then compare prices among options that serve the same purpose. A $379 headband and a $14,000 professional system are designed for different settings.

Fourth: some options require a trained provider

Alongside the devices you operate independently, LENS and QEEG-guided home programs both involve a trained provider — but they use different models.

LENS — provider-administered, passive EEG feedback

A LENS session is passive and typically lasts about three to five minutes. Sensors record EEG activity, and the system returns a very low-energy electromagnetic signal. There is no game, audio interruption or conscious task. LENS is administered by a licensed professional who has completed the required training, or by a trained technician working under that professional’s supervision.

Ochs Labs lists the complete professional system at $13,994, including Foundations training for one person, a configured laptop, software and session equipment. Professional rental is $495 per month with a $2,000 refundable deposit, plus shipping. Those are provider equipment costs; what a client pays per session varies by practice.

LENS may appeal to someone who wants a very brief session or finds active training difficult. It should not be treated as a proven shortcut to peak performance: controlled research is still limited, and the small sham-controlled studies published so far have not shown a significant advantage over sham.

QEEG-guided at-home programs

The structure is similar across providers. Clinical equipment is shipped to your home, and a full-head QEEG is recorded using a gelled cap with nineteen sensors, guided live by video and often paired with a computerized attention test. A clinician reviews the recording, develops the training protocols and revises them over time. Follow-up brain maps may be completed every ten to thirty sessions to measure changes. Training sessions take place at home three or four times a week using a clinical amplifier rather than a headband.

This fills a gap left by the other options. Myndlift includes remote clinician oversight but uses consumer dry-sensor data and does not include a brain map. Most other home systems include neither. Clinician-supervised QEEG programs include both.

Two named examples, with pricing taken from their published pages in September 2026.

Peak Brain Institute — remote programs from $4,999

Ships a full-head EEG cap for the brain map and a two-to-four channel clinical amplifier running EEGer, long-standing clinical neurofeedback software, for the training itself. Nineteen-channel QEEG plus a continuous performance test, reviewed with you rather than handed over as a report. Two months of 25–35 sessions is $4,999 and includes two brain maps with consultations; four months is $7,499, six months $9,499. An at-home QEEG on its own is $1,249. Coaching support seven days a week, and offices in Los Angeles, Orange County, St Louis, New York, London and Stockholm if you would rather map in person and train at home.

Myneurva — at-home programs from $3,999

Nineteen-sensor cap, QEEG read and reviewed by a psychiatrist, and a repeat brain map every ten sessions rather than every thirty. Ten sessions with two QEEGs is $3,999; twenty sessions with three is $5,499; thirty with four is $6,999. A standalone brain scan with a medical review is listed at $999 and currently offered at $699. Sessions run 48 minutes.

What the category buys, and what it costs

The primary difference is the assessment. Most other systems on this page begin training with a general or preselected protocol rather than an individualized measurement. A brain map can also provide an objective baseline and follow-up measure, allowing you to compare changes in the recording with changes you notice subjectively.

Much of the cost reflects clinical time. At $4,000 to $9,500, these programs cost more than the other options on this page, and the outcome depends substantially on the clinician’s interpretation of the data. Ask about the clinician’s experience with your particular concerns before enrolling, and ask what the program does when someone is not responding as expected.

Check two things with any provider in this category. First, "medical-grade" and "FDA-cleared" generally refer to the classification of the EEG amplifier as recording equipment; they do not mean that the program itself has been cleared to treat a condition. Second, confirm that the brain map is repeated during the program rather than performed only at the beginning. Follow-up mapping is what allows the protocol to be reassessed using new data.

How this compares with the target-free approach

These programs and NeurOptimal take different approaches to the same question, and the tradeoffs are worth understanding.

A QEEG program measures brain activity, identifies patterns a clinician believes should change, and trains toward those targets. Its strengths are individualization, a larger research base for protocol training, and measurement at the beginning and end. Its limitation is that the process depends on the quality of the recording and the clinician’s interpretation.

NeurOptimal sets no target and therefore does not require a brain map. It analyses the signal continuously and adapts the feedback as it goes, rather than training against a stored map, which is why no separate mapping session is needed. That also makes it usable by people without professional training, and by several members of a household without individual protocols. The tradeoff is different: there is no targeted protocol, and less peer-reviewed research than protocol-based training has accumulated.

If you want your brain measured and a specialist directing what happens next, a supervised QEEG program is the option that does that, and we do not offer it. If you would rather nothing be aimed at your brain on the strength of someone’s reading of a map, that is the argument for the target-free approach. Both positions are held by serious people.

Device by device

Mendi — $379 (frequently discounted to ~$296)

An entry point that includes an fNIRS headband, phone app and one training game. There is no subscription, while several competing devices have recurring fees. It is rated for ages 5 and up and includes a 60-day money-back guarantee.

Best for: someone curious who wants a focus-training tool rather than a clinical-grade system.

Limits: one sensor region, one exercise, no EEG, one user profile.

Muse S Athena — $474.99 (~$519.98 with a year of Premium)

Muse built its reputation on meditation feedback, and Athena adds fNIRS to the EEG headband. It is designed to be comfortable enough for sleep tracking, which distinguishes it from the other devices in this comparison. Access to the content library requires a subscription after the bundled year.

Best for: meditation and sleep tracking with some brain training layered on.

Limits: it is a wellness device first. Evaluating it against clinical systems compares different product categories.

Narbis — $690

EEG glasses that darken as attention wanders, providing immediate visual feedback without requiring the user to watch a screen. The system includes a preconfigured tablet, so setup is minimal. It uses three EEG channels, is rated for ages 6 and up, includes a 30-day money-back guarantee and has no subscription.

Best for: attention training, particularly for children, where simplicity beats configurability.

Limits: narrow by design; it focuses on attention training.

Myndlift — from $399 for a three-month program

A hybrid model that combines consumer EEG hardware with a remote clinician who sets and adjusts the protocol. Clinician oversight is central to the service, and a professional reviews the data between sessions. The equipment kit starts at $229 plus membership; clinician-guided training is $150/month, with a self-guided plan available afterward for $29/month. A 30-day money-back guarantee is included.

Best for: people who want protocol-based training with professional oversight and no in-person appointments.

Limits: the system is target-driven, with a clinician deciding which patterns to train. Results therefore depend partly on that clinician’s availability and judgment, and ongoing fees should be included in the total cost.

Neurosity Crown — $1,499

An eight-channel dry EEG system designed for developers as well as personal tracking, with an open API. It tracks focus and sleep and allows technically inclined users to build applications using their data. The first year of service is included; after that, membership is $29.99/month or $299.99/year.

Best for: technically inclined users who want data access, not a guided program.

Limits: it is not designed or marketed as a treatment system. Review the return terms before buying: the 90-day guarantee requires at least 30 completed sessions, while the standard return window is 14 days with a 10–15% restocking fee.

Sens.ai — $1,250

The broadest mix of features among the consumer systems compared here: EEG neurofeedback, photobiomodulation and heart-rate-variability training in one headset, with assessments that reselect the protocol. The first 60 days of membership are included. After that, Personal costs $29/month, Duo $39/month and Family $49/month, billed annually. The Family plan supports up to five accounts. Undamaged devices may be returned within 60 days, with return shipping paid by the customer.

Best for: a household wanting one device covering several modalities.

Limits: membership fees resume after 60 days, and a multi-modal system may be less specialized in each area than a device dedicated to one modality. The five-account allowance applies only to the Family plan; Personal supports one user and Duo supports two.

LENS — provider sessions vary; professional system $13,994

LENS is a professional, provider-administered system rather than a home device you buy and operate independently. It measures EEG and briefly returns a very low-energy electromagnetic signal through scalp sensors. Sessions are passive and typically take about three to five minutes.

Best for: someone who wants a brief, passive form of EEG feedback with a trained provider directing the process.

Limits: access and client pricing depend on the provider; it is not designed for independent household use. Independent controlled evidence is limited, and published small trials have not shown a significant advantage over sham.

NeurOptimal — rent $650–799/month, or buy $5,995–9,995

Because we rent NeurOptimal, its relevant strengths and limitations are stated directly below.

NeurOptimal is a professional-grade non-linear dynamical system — the same equipment and specifications used by clinicians. It sets no targets and makes no diagnoses; instead, it responds to changes in brain activity as they occur. Because no individualized protocol needs to be configured, the system can be used by people without professional training. A purchased system has no per-user limit, and our Elite rental supports up to six people.

The primary barrier is the purchase price: $5,995 to $9,995, substantially more than the consumer devices above. Renting provides access to the system without requiring an outright purchase.

Best for: whole-household use, people who want clinical-grade equipment without buying it, anyone who would rather not manage protocols.

Limits: the hardware is bulkier than a headband, buying it outright is a serious expense, and renting is time-limited by design.

Rent, lease, or buy: the arithmetic

Most neurofeedback is completed in a defined series of sessions over several months rather than used indefinitely. This changes the cost comparison.

Our rental pricing (28-day month):

Shipping is $35 each way.

Lease pricing: $699/mo for 3 months, $599/mo for 6 months, $499/mo for 12 months. See lease options.

A typical two-to-four month program runs $1,400–3,300. See programs and costs.

Here is how the costs compare:

• Against buying NeurOptimal: purchase prices range from $5,995 to $9,995, while four months of Elite rental costs $3,196. Buying may make more sense if you expect to use the system for years; renting is designed for a defined program with an end date.

• Against a consumer device: two months of Elite rental costs $1,598, while Mendi costs $379 and remains yours. If Mendi meets your needs, purchasing it may be the more practical choice. Renting NeurOptimal becomes relevant when you want professional-grade equipment, training for several household members, or a system that does not require protocol configuration.

• Against in-office sessions: our NYC office is $150 per session, or $1,150 for ten. Twenty sessions in-office is $2,300 plus twenty round trips. Twenty sessions at home on an Elite rental is $799 plus $70 shipping.

• Against subscription systems: four months of Myndlift clinician-guided training is $600 in membership plus an equipment kit from $229. Sens.ai is $1,250 with the first 60 days of membership included. Four months of Elite rental is $3,196. What the money buys differs in each case: a clinician reviewing your data, a device you keep, or professional-grade equipment for a household with nothing to configure.

• Against a supervised QEEG program: two months with Peak Brain is $4,999 and includes two brain maps and 25–35 sessions with a clinician setting the protocols. Ten sessions with Myneurva plus two brain maps is $3,999. Two months of Elite rental is $1,598 plus $70 shipping, with no map, no clinician and no protocol. These buy different things, and which one is right turns on whether you want your brain measured and someone directing what happens next.

• Against provider-administered LENS: client pricing varies by practice, so there is no standard program total to compare. The $13,994 purchase price and $495 monthly rental are professional equipment costs, and access requires a trained licensed provider or supervised technician.

Choosing, by budget and situation

The quick-glance table above starts from what you are dealing with. This one starts from what you are willing to spend and how many people are involved. Use whichever is the tighter constraint.

• Under $500, one user, focus training → Mendi, or Narbis if it is for a child.

• Meditation and sleep first → Muse S Athena.

• Want a clinician involved → Myndlift.

• Want a very brief, passive session administered by a trained provider → LENS.

• Want the data and can code → Neurosity Crown.

• One device, several modalities, $1,250 → Sens.ai.

• Whole household, clinical-grade, defined program → rent NeurOptimal.

• Whole household, clinical-grade, indefinite use → buy NeurOptimal.

• Want your brain mapped and a clinician directing the training → a supervised QEEG program such as Peak Brain or Myneurva. We do not offer this category.

Common questions

Which home neurofeedback device is best?

There is no single answer, because the four categories on this page are built for different things. A device that trains attention with one protocol is not a worse version of a supervised clinical program; it is a different product, at a different price, for a different situation. Start with the quick-glance table, decide which category fits what brought you here, and compare prices within that category rather than across all of them.

Is BrainTap neurofeedback?

No. BrainTap is an audio-visual entrainment device. It plays flashing light and pulsing tones at a target frequency, but it does not measure brain activity and does not change what it plays based on what your brain is doing. Neurofeedback requires that loop — measurement, then feedback derived from the measurement. BrainTap sells for $797, or $907 with a year of app membership, and $29.99 a month or $260 a year after that. Compare it with other entrainment and relaxation tools rather than with the EEG and fNIRS systems on this page.

Is a $400 device as good as a $6,000 one?

Within its intended use, possibly. A consumer fNIRS headband trains prefrontal blood flow, but it is not doing what a professional EEG system does. Compare the technology and intended use before comparing price.

How many sessions before anything happens?

Practitioner guidance commonly puts a course at 20 to 40 sessions over two to four months. Claims of dependable results within five sessions should be viewed cautiously. Ochs Labs says LENS clients average roughly nine to eleven sessions, but that is a manufacturer estimate, not a dependable outcome timetable.

Do subscriptions ever end?

Depends on the device. Mendi and Narbis have none. Muse, Sens.ai, Myndlift and Neurosity all have recurring costs after an initial period: Sens.ai from $29/month after 60 days, Neurosity $29.99/month after the first year, Myndlift $150/month with a clinician or $29/month self-guided. Add the ongoing cost across your intended program length before comparing sticker prices — it frequently reverses the order.

What should I ask that product pages do not answer?

Ask three questions. Are the sensors dry or applied with a conductive medium, and is the connection wired or Bluetooth? Does the system train toward a preset target or give the brain information without one? How much of the frequency range does it monitor? Together, these answers clarify what the device is designed to do.

Do I need a brain map first?

Protocol-based neurofeedback generally requires a QEEG brain map to identify the patterns a clinician will target and set the training protocols. NeurOptimal does not require a separate map because it analyses brainwave activity and adapts the feedback in real time rather than training toward predetermined targets. LENS also uses EEG readings during the process to guide the provider.

You can still have a brain map as a baseline for future comparison. The more important question is what you plan to do with the information: use it to set targeted protocols, establish an objective baseline, or provide additional clinical guidance.

Can more than one person use one device?

Mendi, Narbis, Muse and Neurosity are effectively single-user. Sens.ai supports one account on Personal, two on Duo and five on Family. NeurOptimal supports up to six people on an Elite rental and has no per-user limit when purchased. LENS is provider equipment and is not sold as an independent multi-user home device.

Is home neurofeedback safe?

The consumer systems here are non-invasive and do not electrically stimulate the brain. LENS is the exception on this page: it returns a very low-energy electromagnetic signal through the scalp sensors, and it is administered by a trained provider rather than used independently at home. Some people report transient tiredness or headache after neurofeedback. Anyone with a seizure disorder, or on medication that may need adjusting, should involve their physician — and no device or provider session replaces medical care.